{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T06:22:09Z","timestamp":1778048529548,"version":"3.51.4"},"reference-count":40,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,3,6]],"date-time":"2026-03-06T00:00:00Z","timestamp":1772755200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,3,6]],"date-time":"2026-03-06T00:00:00Z","timestamp":1772755200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,3,6]]},"DOI":"10.1109\/wacv61042.2026.00290","type":"proceedings-article","created":{"date-parts":[[2026,5,5]],"date-time":"2026-05-05T19:59:32Z","timestamp":1778011172000},"page":"2964-2973","source":"Crossref","is-referenced-by-count":0,"title":["Scalpel: Fine-Grained Alignment of Attention Activation Manifolds via Mixture Gaussian Bridges to Mitigate Multimodal Hallucination"],"prefix":"10.1109","author":[{"given":"Ziqiang","family":"Shi","sequence":"first","affiliation":[{"name":"Fujitsu Research & Development Center Co.,LTD.,Beijing,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rujie","family":"Liu","sequence":"additional","affiliation":[{"name":"Fujitsu Research & Development Center Co.,LTD.,Beijing,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shanshan","family":"Yu","sequence":"additional","affiliation":[{"name":"Fujitsu Limited,Tokyo,Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Satoshi","family":"Munakata","sequence":"additional","affiliation":[{"name":"Fujitsu Limited,Tokyo,Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Koichi","family":"Shirahata","sequence":"additional","affiliation":[{"name":"Fujitsu Limited,Tokyo,Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"issue":"1","key":"ref1","doi-asserted-by":"crossref","first-page":"72","DOI":"10.1109\/89.365379","volume":"3","year":"1995","journal-title":"IEEE transactions on speech and audio processing"},{"key":"ref2","article-title":"Qwen2. 5-vl technical report","author":"Bai","year":"2025"},{"key":"ref3","first-page":"5802","article-title":"The schr\u00f6dinger bridge between gaussian measures has a closed form","volume-title":"International Conference on Artificial Intelligence and Statistics","author":"Bunne"},{"key":"ref4","article-title":"Discovering latent knowledge in language models without supervision","author":"Burns","year":"2022"},{"key":"ref5","article-title":"Perturbollava: Reducing multimodal hallucinations with perturbative visual training","author":"Chen","year":"2025"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.00398"},{"issue":"3","key":"ref7","first-page":"6","volume":"2","author":"Chiang","year":"2023","journal-title":"Vicuna: An open-source chatbot impressing gpt-4 with 90%* chatgpt quality"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01356"},{"key":"ref9","article-title":"Mme: A comprehensive evaluation benchmark for multimodal large language models","author":"Fu","year":"2023"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01274"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00686"},{"key":"ref12","article-title":"Language models (mostly) know what they know","author":"Kadavath","year":"2022"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1007\/BF01457949"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01316"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.3934\/dcds.2014.34.1533"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1797"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.20"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref20","article-title":"Mia-dpo: Multi-image augmented direct preference optimization for large vision-language models","author":"Liu","year":"2024"},{"key":"ref21","article-title":"Dama: Data-and model-aware alignment of multi-modal llms","author":"Lu","year":"2025"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1561\/2200000073"},{"key":"ref23","article-title":"Go with the flow: Fast diffusion for gaussian mixture models","author":"Rapakoulias","year":"2024"},{"key":"ref24","author":"Saunders","year":"2022","journal-title":"Self-critiquing models for assisting human evaluators"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20074-8_9"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.findings-acl.48"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-acl.775"},{"key":"ref28","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-025-02607-z"},{"issue":"11","key":"ref30","article-title":"Visualizing data using t-sne","volume":"9","author":"Van der Maaten","year":"2008","journal-title":"Journal of machine learning research"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref32","article-title":"Language models are open knowledge graphs","author":"Wang","year":"2020"},{"key":"ref33","article-title":"Qwen2-vl: Enhancing vision-language model\u2019s perception of the world at any resolution","author":"Wang","year":"2024"},{"key":"ref34","article-title":"Eaco: Enhancing alignment in multimodal llms via critical observation","author":"Wang","year":"2024"},{"key":"ref35","article-title":"Don\u2019t miss the forest for the trees: Attentional vision calibration for large vision language models","author":"Woo","year":"2024"},{"key":"ref36","article-title":"Mmada: Multimodal large diffusion language models","author":"Yang","year":"2025"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.01363"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0823"},{"key":"ref39","article-title":"Seeing clearly by layer two: Enhancing attention heads to alleviate hallucination in lvlms","author":"Zhang","year":"2024"},{"key":"ref40","article-title":"Minigpt-4: Enhancing vision-language understanding with advanced large language models","author":"Zhu","year":"2023"}],"event":{"name":"2026 IEEE\/CVF Winter Conference on Applications of Computer Vision (WACV)","location":"Tucson, AZ, USA","start":{"date-parts":[[2026,3,6]]},"end":{"date-parts":[[2026,3,10]]}},"container-title":["2026 IEEE\/CVF Winter Conference on Applications of Computer Vision (WACV)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11491838\/11491925\/11492144.pdf?arnumber=11492144","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T06:09:16Z","timestamp":1778047756000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11492144\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,6]]},"references-count":40,"URL":"https:\/\/doi.org\/10.1109\/wacv61042.2026.00290","relation":{},"subject":[],"published":{"date-parts":[[2026,3,6]]}}}